PROJECT SUMMARY Obesity is one of the nation?s most pressing public health concerns and sugar-sweetened beverages (SSBs) are a significant contributor to adult and childhood obesity. To try to reduce SSB intake, bills have been introduced in U.S. states and cities to place health warning labels on SSB containers and/or advertisements. However, there is little empirical data on how such labels influence consumers. Research on tobacco and nutrition labels suggests labels can positively impact behavior, but the label design matters. The primary objective of this proposal is to determine, before wide-scale implementation, to what degree SSB warning labels increase consumers? knowledge about the potential health harms of SSBs and reduce SSB intake. The studies are designed to answer three additional questions: 1) Do some warning labels work better than others? 2) What is the effect of warning labels over time? 3) If warning labels influence behavior, is it because they increase knowledge or simply provide a salient reminder that some drinks are less healthy? Our first study aim is to compare the effects of SSB warning labels on parents? SSB-related beliefs and purchase intentions. We will recruit a nationally representative sample of 1,000 racially and ethnically diverse parents of a child under 12 years old to participate in a randomized, controlled online experiment during which parents will shop in a virtual convenience store. Parents will be randomized to 1 of 4 different approaches to labeling: 1) calorie labels (control); 2) text warning labels (e.g. warning: drinking beverages with added sugars contributes to obesity, diabetes, and tooth decay); 3) graphic warning labels displaying amounts of sugar in beverages; or 4) graphic warning labels displaying negative health consequences associated with overconsumption of SSBs. The second aim is to test the effect of warning labels on total kilocalories purchased and consumed by parents and children. We will recruit 405 racially and ethnically diverse parent-child pairs to participate in a randomized, controlled lab-based eating behavior study that will capture beverage and snack purchases from a snack shop and measure how much participants consume while watching a television show. Participants will be randomized to 1 of 3 conditions: 1) calorie labels (control); 2) text warning labels; or 3) graphic warning labels displaying either amounts of sugar or negative health consequences based on the most effective labeling strategy in study one. The third aim is to test the effect of repeated exposure to warning labels on total kilocalories purchased over time and assess whether knowledge or salience better explain label effects. In this study, we will recruit 450 racially and ethnically diverse parents to participate in a randomized, controlled experiment that involves buying snacks and beverages for six weeks via an online store that ships participants their purchases. Participants will be randomized to one of the same three label conditions described for the second study. Labels will be removed during the last two weeks to determine whether education or salience plays a greater role in driving behavior change.